forgeNet: a graph deep neural network model using tree-based ensemble classifiers for feature graph construction.

MOTIVATION: A unique challenge in predictive model building for omics data has been the small number of samples (n) versus the large amount of features (p). This 'n≪p' property brings difficulties for disease outcome classification using deep learning techniques. Sparse learning by incorporating known functional relationships between the biological units, such as the graph-embedded deep feedforward network (GEDFN) model, has been a solution to this issue. However, such methods require an existing feature graph, and potential mis-specification of the feature graph can be harmful on classification and feature selection. RESULTS: To address this limitation and develop a robust classification model without relying on external knowledge, we propose a forest graph-embedded deep feedforward network (forgeNet) model, to integrate the GEDFN architecture with a forest feature graph extractor, so that the feature graph can be learned in a supervised manner and specifically constructed for a given prediction task. To validate the method's capability, we experimented the forgeNet model with both synthetic and real datasets. The resulting high classification accuracy suggests that the method is a valuable addition to sparse deep learning models for omics data. AVAILABILITY AND IMPLEMENTATION: The method is available at https://github.com/yunchuankong/forgeNet. CONTACT: tianwei.yu@emory.edu. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

DNLC: differential network local consistency analysis.

BACKGROUND: The biological network is highly dynamic. Functional relations between genes can be activated or deactivated depending on the biological conditions. On the genome-scale network, subnetworks that gain or lose local expression consistency may shed light on the regulatory mechanisms related to the changing biological conditions, such as disease status or tissue developmental stages. RESULTS: In this study, we develop a new method to select genes and modules on the existing biological network, in which local expression consistency changes significantly between clinical conditions. The method is called DNLC: Differential Network Local Consistency. In simulations, our algorithm detected artificially created local consistency changes effectively. We applied the method on two publicly available datasets, and the method detected novel genes and network modules that were biologically plausible. CONCLUSIONS: The new method is effective in finding modules in which the gene expression consistency change between clinical conditions. It is a useful tool that complements traditional differential expression analyses to make discoveries from gene expression data. The R package is available at https://cran.r-project.org/web/packages/DNLC.

A hypergraph-based method for large-scale dynamic correlation study at the transcriptomic scale.

BACKGROUND: The biological regulatory system is highly dynamic. Correlations between functionally related genes change over different biological conditions, which are often unobserved in the data. At the gene level, the dynamic correlations result in three-way gene interactions involving a pair of genes that change correlation, and a third gene that reflects the underlying cellular conditions. This type of ternary relation can be quantified by the Liquid Association statistic. Studying these three-way interactions at the gene triplet level have revealed important regulatory mechanisms in the biological system. Currently, due to the extremely large amount of possible combinations of triplets within a high-throughput gene expression dataset, no method is available to examine the ternary relationship at the biological system level and formally address the false discovery issue. RESULTS: Here we propose a new method, Hypergraph for Dynamic Correlation (HDC), to construct module-level three-way interaction networks. The method is able to present integrative uniform hypergraphs to reflect the global dynamic correlation pattern in the biological system, providing guidance to down-stream gene triplet-level analyses. To validate the method's ability, we conducted two real data experiments using a melanoma RNA-seq dataset from The Cancer Genome Atlas (TCGA) and a yeast cell cycle dataset. The resulting hypergraphs are clearly biologically plausible, and suggest novel relations relevant to the biological conditions in the data. CONCLUSIONS: We believe the new approach provides a valuable alternative method to analyze omics data that can extract higher order structures. The software is at https://github.com/yunchuankong/HypergraphDynamicCorrelation .

A graph-embedded deep feedforward network for disease outcome classification and feature selection using gene expression data.

Motivation: Gene expression data represents a unique challenge in predictive model building, because of the small number of samples (n) compared with the huge amount of features (p). This 'n≪p' property has hampered application of deep learning techniques for disease outcome classification. Sparse learning by incorporating external gene network information could be a potential solution to this issue. Still, the problem is very challenging because (i) there are tens of thousands of features and only hundreds of training samples, (ii) the scale-free structure of the gene network is unfriendly to the setup of convolutional neural networks. Results: To address these issues and build a robust classification model, we propose the Graph-Embedded Deep Feedforward Networks (GEDFN), to integrate external relational information of features into the deep neural network architecture. The method is able to achieve sparse connection between network layers to prevent overfitting. To validate the method's capability, we conducted both simulation experiments and real data analysis using a breast invasive carcinoma RNA-seq dataset and a kidney renal clear cell carcinoma RNA-seq dataset from The Cancer Genome Atlas. The resulting high classification accuracy and easily interpretable feature selection results suggest the method is a useful addition to the current graph-guided classification models and feature selection procedures. Availability and implementation: The method is available at https://github.com/yunchuankong/GEDFN. Supplementary information: Supplementary data are available at Bioinformatics online.

Stable quantum key distribution using a silicon photonic transceiver.

The wider deployment of commercial quantum key distribution (QKD) may benefit from an integrated system with reduced cost, small form-of-factor and high robutness. Silicon photonic circuits are good candidates while their performance stability in some contexts remains a challenge. We demonstrate a silicon photonic QKD transceiver based on time-bin protocol. The stability of the transceiver is investigated and a feedback function is proposed to improve the temperature-dependent performance of the transceiver. With the help of a faster data-processing ability, such scheme can facilitate more application scenarios, therefore achieving wider implementation of QKD in the future.

High repetition rate correlated photon pair generation in integrated silicon nanowires.

Integrated single-photon sources are a key component for photonic quantum technology but are generally limited to low single-photon rates. For sources based on photon pair generation by four-wave mixing, increasing the repetition rate of pump laser pulses is a straightforward way to enhance the single-photon rate, but the benefits and practical limitations have not yet been demonstrated and analyzed in a CMOS-compatible platform. In this work, we demonstrate correlated photon pair generation in integrated silicon nanowires and systematically analyze the count rate and coincidence to accidental ratio as the pump rate is varied between 156.25 MHz and 10 GHz. We show that the highest useful pump rate is limited by the timing resolution of the single-photon detection system, and that in this regime, the nonlinear loss of the silicon nanowire does not have a significant effect on the single-photon generation.

Immunologic mechanisms of seasonal influenza vaccination administered by microneedle patch from a randomized phase I trial.

In a phase 1 randomized, single-center clinical trial, inactivated influenza virus vaccine delivered through dissolvable microneedle patches (MNPs) was found to be safe and immunogenic. Here, we compare the humoral and cellular immunologic responses in a subset of participants receiving influenza vaccination by MNP to the intramuscular (IM) route of administration. We collected serum, plasma, and peripheral blood mononuclear cells in 22 participants up to 180 days post-vaccination. Hemagglutination inhibition (HAI) titers and antibody avidity were similar after MNP and IM vaccination, even though MNP vaccination used a lower antigen dose. MNPs generated higher neuraminidase inhibition (NAI) titers for all three influenza virus vaccine strains tested and triggered a larger percentage of circulating T follicular helper cells (CD4 + CXCR5 + CXCR3 + ICOS + PD-1+) compared to the IM route. Our study indicates that inactivated influenza virus vaccination by MNP produces humoral and cellular immune response that are similar or greater than IM vaccination.

A Deep Neural Network Model using Random Forest to Extract Feature Representation for Gene Expression Data Classification.

In predictive model development, gene expression data is associated with the unique challenge that the number of samples (n) is much smaller than the amount of features (p). This "n ≪ p" property has prevented classification of gene expression data from deep learning techniques, which have been proved powerful under "n > p" scenarios in other application fields, such as image classification. Further, the sparsity of effective features with unknown correlation structures in gene expression profiles brings more challenges for classification tasks. To tackle these problems, we propose a newly developed classifier named Forest Deep Neural Network (fDNN), to integrate the deep neural network architecture with a supervised forest feature detector. Using this built-in feature detector, the method is able to learn sparse feature representations and feed the representations into a neural network to mitigate the overfitting problem. Simulation experiments and real data analyses using two RNA-seq expression datasets are conducted to evaluate fDNN's capability. The method is demonstrated a useful addition to current predictive models with better classification performance and more meaningful selected features compared to ordinary random forests and deep neural networks.